Báo cáo khoa học: "Exploiting Feature Hierarchy for Transfer Learning in Named Entity Recognition" ppt

... Proceedings of ACL-08: HLT, pages 245–253, Columbus, Ohio, USA, June 2008. c 2008 Association for Computational Linguistics Exploiting Feature Hierarchy for Transfer Learning in Named Entity Recognition Andrew ... transfer learning in named entity recognition, motivated by the common structure of feature spaces for this task across natural language data sets....

Ngày tải lên: 23/03/2014, 17:20

... s 1 ). For non-binary features, we first calculate real values and then binarize them into 10 quantile bins. Feature group URL. This group exploits NE information in URLs. The feature URL-SUBPART (line ... search engine and use similar contexts in the search results as additional information for correctly classifying the token. We achieve strong gains in NER performance on news,...

Ngày tải lên: 23/03/2014, 16:20

... rel- atively simple starting point in order to gain experience in implementing useful semantic markup in a publishing workflow without a substantial word-sense disambigua- tion effort. GO terms are ... Lucinda, outlined in Table 1, for generating the input for the recogniser from external data. We use the plain-text OBO 1.2 format, which is the definitive format for the disseminat...

Ngày tải lên: 31/03/2014, 01:20

... number of training examples in each sense in such a new training set is the same as that in the official train- ing data set of w. A WSD classifier was then trained on this new training set, and ... correct sense, accord- ing to some existing sense inventory in a diction- ary. This annotated corpus then serves as the training material for a learning algorithm. After training...

Ngày tải lên: 08/03/2014, 04:22

... Substituting the probabilities in the PMI for- mula with the previously introduced Web statistics, we obtain: Qsp Asp Qsp Asp Maximal Likelihood Ratio (MLHR) is also used for word co-occurrence mining ... performance as proof for the difference between our task and classic co- occurrence mining. Another indication for this is the fact that MLHR and PMI performances are compa- rable,...

Ngày tải lên: 08/03/2014, 07:20

... Ting Liu, Wen- liang Chen, and Haizhou Li. 2011. Joint models for chinese pos tagging and dependency parsing. In EMNLP 2011, pages 1180–1191. Ting Liu, Jinshan Ma, and Sheng Li. 2006. Building a ... Online large-margin training of dependency parsers. In Proceedings of ACL 2005, pages 91–98. Zheng-Yu Niu, Haifeng Wang, and Hua Wu. 2009. Ex- ploiting heterogeneous treebanks for parsing....

Ngày tải lên: 16/03/2014, 19:20

... tried the corpus weighting method when combining CDT P S with CTB training set (abbre- viated as CTB for simplicity) as training data, by gradually increasing the weight (including 1, 2, 5, 10, 20, ... 60k Chinese sentences, anno- tated with POS tag information and dependency structure information (including 28 POS tags, and 24 dependency tags) (Liu et al., 2006). We did not use POS tag inf...

Ngày tải lên: 17/03/2014, 01:20

... techniques for selecting an appropriate content according to contextual interpretation [6], and on the way of presenting explanations by taking the information Seeking person's knowledge into ... concentrated our efforts on presenting different types of knowledge and their interrelations because this kind of infor- mation is typically relevant for explanations. We formally re...

Ngày tải lên: 24/03/2014, 05:21

... labelled information for training. Our sys- tem extracts transliteration pairs in an unsupervised fashion. It is also able to utilize labelled information if available, obtaining improved performance. We ... supervised/semi-supervised infor- mation for learning. We propose a flexible genera- tive model for transliteration mining usable for both unsupervised and semi-supervised lear...

Ngày tải lên: 19/02/2014, 19:20

... on dictionary, machine translation or language learning, losing out on synergy that can reduce inefficiencies in the user experience. This paper presents Engkoo, a system for explor- ing and learning language. ... versa. Finally, the mined bilingual term/sentence pairs, together with their parsed in- formation, are stored and indexed with a multi-level indexing engine, a core compone...

Ngày tải lên: 20/02/2014, 05:20